{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.06988608568034105,
  "eval_steps": 500,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 1.1361215114593506,
      "learning_rate": 3.4935718278367804e-05,
      "loss": 7.3406,
      "step": 500
    },
    {
      "epoch": 0.01,
      "eval_loss": 5.7022929191589355,
      "eval_runtime": 29.28,
      "eval_samples_per_second": 27.083,
      "eval_steps_per_second": 3.415,
      "step": 500
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.8447604775428772,
      "learning_rate": 6.987143655673561e-05,
      "loss": 5.4549,
      "step": 1000
    },
    {
      "epoch": 0.01,
      "eval_loss": 5.279269695281982,
      "eval_runtime": 29.2904,
      "eval_samples_per_second": 27.074,
      "eval_steps_per_second": 3.414,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.7530624866485596,
      "learning_rate": 0.00010480715483510341,
      "loss": 5.2047,
      "step": 1500
    },
    {
      "epoch": 0.02,
      "eval_loss": 5.1141037940979,
      "eval_runtime": 29.315,
      "eval_samples_per_second": 27.051,
      "eval_steps_per_second": 3.411,
      "step": 1500
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.9565019607543945,
      "learning_rate": 0.00013974287311347122,
      "loss": 5.003,
      "step": 2000
    },
    {
      "epoch": 0.03,
      "eval_loss": 4.823694229125977,
      "eval_runtime": 29.3039,
      "eval_samples_per_second": 27.061,
      "eval_steps_per_second": 3.413,
      "step": 2000
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.7762724161148071,
      "learning_rate": 0.00017467859139183903,
      "loss": 4.7353,
      "step": 2500
    },
    {
      "epoch": 0.03,
      "eval_loss": 4.599145412445068,
      "eval_runtime": 29.2873,
      "eval_samples_per_second": 27.077,
      "eval_steps_per_second": 3.414,
      "step": 2500
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.6923783421516418,
      "learning_rate": 0.00020961430967020682,
      "loss": 4.565,
      "step": 3000
    },
    {
      "epoch": 0.04,
      "eval_loss": 4.469151020050049,
      "eval_runtime": 29.3283,
      "eval_samples_per_second": 27.039,
      "eval_steps_per_second": 3.41,
      "step": 3000
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.6830642819404602,
      "learning_rate": 0.0002445500279485746,
      "loss": 4.4583,
      "step": 3500
    },
    {
      "epoch": 0.05,
      "eval_loss": 4.370917797088623,
      "eval_runtime": 29.2477,
      "eval_samples_per_second": 27.113,
      "eval_steps_per_second": 3.419,
      "step": 3500
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.6420345902442932,
      "learning_rate": 0.0002499762209394527,
      "loss": 4.3647,
      "step": 4000
    },
    {
      "epoch": 0.06,
      "eval_loss": 4.299973011016846,
      "eval_runtime": 29.2476,
      "eval_samples_per_second": 27.113,
      "eval_steps_per_second": 3.419,
      "step": 4000
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.6263813972473145,
      "learning_rate": 0.00024988650423117477,
      "loss": 4.3035,
      "step": 4500
    },
    {
      "epoch": 0.06,
      "eval_loss": 4.233543872833252,
      "eval_runtime": 29.222,
      "eval_samples_per_second": 27.137,
      "eval_steps_per_second": 3.422,
      "step": 4500
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.6064193248748779,
      "learning_rate": 0.0002497300853885875,
      "loss": 4.2468,
      "step": 5000
    },
    {
      "epoch": 0.07,
      "eval_loss": 4.18317985534668,
      "eval_runtime": 29.2822,
      "eval_samples_per_second": 27.081,
      "eval_steps_per_second": 3.415,
      "step": 5000
    }
  ],
  "logging_steps": 500,
  "max_steps": 71545,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5000,
  "total_flos": 2.090336256e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}